Evaluation of Diagnostic Performance of Machine Learning Algorithms to Classify the Fetal Heart Rate Baseline From Cardiotocograph

Cardiotocography (CTG) is the widely used cost-effective, non-invasive technique to monitor fetal heart and mother’s uterine contraction pressure assess wellbeing of fetus. The most important parameters baseline upon which other viz. acceleration, deceleration variability depend. Accurate classification into either normal, bradycardia or tachycardia thus fetal-health. Since visual estimation has its limitations, authors use various Machine Learning Algorithms classify baseline. 110 CTG traces from CTU-UHB dataset, were divided three subsets using stratified sampling ensure that sample accurate depiction population. results analyzed statistical methods compared with by obstetricians. FURIA provided greatest accuracy 98.11%. From analysis Bland-Altman Plot was also found have best agreement physicians’ estimation.